From 3aefb3172756e6648cef49a53cd9615265c5ce0e Mon Sep 17 00:00:00 2001 From: Hongsheng Liu Date: Wed, 12 Aug 2026 12:01:32 +0800 Subject: [PATCH] docs(diffusion): remove VAE policy attribution Signed-off-by: Hongsheng Liu --- docs/user_guide/diffusion/parallelism/vae_parallelism.md | 2 -- 1 file changed, 2 deletions(-) diff --git a/docs/user_guide/diffusion/parallelism/vae_parallelism.md b/docs/user_guide/diffusion/parallelism/vae_parallelism.md index 08322f486..bb3aca513 100644 --- a/docs/user_guide/diffusion/parallelism/vae_parallelism.md +++ b/docs/user_guide/diffusion/parallelism/vae_parallelism.md @@ -161,8 +161,6 @@ vllm serve Wan-AI/Wan2.1-T2V-1.3B-Diffusers --omni \ - It requires `vae_patch_parallel_size` to **match the DiT process group size**. `auto` quietly falls back to tile decode for a partial group; an explicitly forced spatial mode logs a warning. - The decoder's context-aware wrappers retain the direct path, so one process can alternate small/large and portrait/landscape requests safely. -The automatic policy is adapted from [SGLang #28071](https://github.com/sgl-project/sglang/pull/28071) and narrowed to vLLM-Omni's parity-tested Wan backend. - For end-to-end latency/throughput, launch serving with the desired `vae_parallel_mode` and use the existing diffusion serving benchmark: ```bash